Research map / Optimization / Nonconvex and smooth optimization
A Hybrid Variance-Reduced Method for Decentralized Stochastic Non-Convex Optimization
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| combine local stochastic gradients with SARAH-style recursive variance reduction and global gradient tracking | decentralized online non-convex optimization with SFO · a9a, covertype, KDD98, and MiniBooNE | loss versus epochs | Helped |
| use the hybrid estimator with an interior weight, β∈(0,1), rather than β=1 (ablation) | GT-HSGD · Decentralized online non-convex optimization with SFO | oracle complexity | Helped |
| use the hybrid estimator with an interior weight, β∈(0,1), rather than β=0 (ablation) | GT-HSGD · Decentralized online non-convex optimization with SFO | oracle complexity | Helped |
| use one stochastic gradient query per update instead of double-loop methods with large minibatches | decentralized online variance-reduced optimization · Online SFO implementation | implementation practicality | Helped |
| use GT-HSGD in the sufficiently small error-tolerance regime | decentralized online optimization · Small ε regime | oracle complexity per node: O(n^{-1}ε^{-3}) | Helped |
| reuse parameters tuned on the complete graph across other tested network topologies | GT-HSGD · a9a and covertype | convergence rate | No clear effect |
Closest papers by meaning
- Decentralized Convex Finite-Sum Optimization with Better Dependence on Condition Numbers (ICML 2024 · similarity 0.92)
- Jointly Improving the Sample and Communication Complexities in Decentralized Stochastic Minimax Optimization (AAAI 2024 · similarity 0.91)
- An Online Optimization Perspective on First-Order and Zero-Order Decentralized Nonsmooth Nonconvex Stochastic Optimization (ICML 2024 · similarity 0.90)
- Decentralized Gradient-Free Methods for Stochastic Non-smooth Non-convex Optimization (AAAI 2024 · similarity 0.90)
- Stability and Generalization of Zeroth-Order Decentralized Stochastic Gradient Descent with Changing Topology (AAAI 2025 · similarity 0.89)
- Enhancing Parallelism in Decentralized Stochastic Convex Optimization (ICML 2025 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.